sign in sign up
<<
Home , EPPTB, Octopamine, Phenethylamine, Phenylephrine, Prazosin, Synephrine

www.nature.com/scientificreports

OPEN Diferential mechanisms of action of the trace , and on the Received: 15 March 2019 Accepted: 29 June 2019 porcine coronary and mesenteric Published: xx xx xxxx Andy Hsien Wei Koh 1, Russ Chess-Williams1,2 & Anna Elizabeth Lohning1

Trace amines such as p-tyramine, p-octopamine and p-synephrine are found in low concentrations in and . Consumption of pre-workout supplements containing these -derived amines has been associated with cardiovascular side efects. The aim of this study was to determine the mechanisms of action of these trace amines on porcine isolated coronary and mesenteric . Noradrenaline caused contraction of mesenteric arteries and relaxation of coronary arteries. In both tissues, all three trace amines induced contractions with similar potencies and responses were unafected by the β-adrenoceptor antagonist (1 µM), the synthase inhibitor L-NNA (100 µM), or the TAAR-1 antagonist, EPPTB (100 nM). However, the contractile responses of mesenteric arteries, but not coronary arteries, were signifcantly reduced by depletion of endogenous noradrenaline. Mesenteric responses to all three amines were abolished in the presence of (1 µM) whereas residual contractile responses remained in the coronary artery which were inhibited by a high concentration (100 µM) of EPPTB. The results suggest complex responses of the coronary artery to the trace amines, with activity at α1-adrenoceptors and potentially TAARs other than TAAR-1. In contrast the actions of the amines on the mesenteric artery appeared to involve indirect

sympathomimetic actions and direct actions on α1-adrenoceptors.

p-Tyramine (tyramine), p-octopamine (octopamine) and p-synephrine (synephrine) are substituted phenethyl- amines with a phenolic hydroxyl group in the para-position (Fig. 1). Tey are found in nanomolar concentrations in the mammalian nervous system and have thus been described as “trace amines”1. Tey also occur naturally in plants such as Citrus aurantium2. Since 2004, Citrus aurantium extracts have been marketed as ergogenic and weight-loss aids, but there are limited studies to support this claim3. Te safety of Citrus aurantium-listed supplements is still debated as adverse cardiovascular efects have been associated with their use4. A number of actions of these trace amines on cardiovascular tissues has been reported. In some vascular tissues, tyramine has been shown to act as an indirectly acting sympathomimetic agent, promoting the release of endogenous noradrenaline5. Moreover, tyramine and octopamine cause nitric oxide dependant vasodilatory responses in pre-contracted rat mesenteric vascular beds6. Synephrine and octopamine are weak direct 7–9 efects at the α1-adrenoceptors of the isolated rat . In addition, octopamine and synephrine have been shown to exert weak direct β1-adrenoceptor agonist efects on isolated cardiac tissues and to activate cloned 10,11 β2-adrenoceptors . Tus, the amines can potentially exert cardiovascular efects via both direct or indirect mechanisms, but the extent of these actions in diferent vessels of the same species has not been explored. Recently it has been recognised that these amines may exhibit some activity mediated via trace -associated receptors (TAARs)12,13. Six subtypes of this have been identifed, but knowledge of their distribution and functions is limited. All three amines have been shown to activate human TAAR-1 expressed in cloned cells and increase cAMP levels, with tyramine being the most potent drug12–14. It has been hypothesised

1Faculty of Health Sciences and Medicine, Bond University, 4229, Queensland, Australia. 2Centre for Urology Research, Faculty of Health Sciences and Medicine, Bond University, 4229, Queensland, Australia. Correspondence and requests for materials should be addressed to A.E.L. (email: [email protected])

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 1 www.nature.com/scientificreports/ www.nature.com/scientificreports

OH H NH2 N

HO HO Tyramine Synephrine

OH OH

NH2 HO NH2

HO HO Octopamine Noradrenaline

Figure 1. Chemical structures of the trace amines (tyramine, octopamine, and synephrine) and noradrenaline.

Figure 2. Initial cumulative-dose responses of trace amines and noradrenaline on isolated mesenteric artery. Te amines shown are tyramine (● n = 6), synephrine (■, n = 6) and octopamine (▲, n = 6). Te data are presented as percent of contraction to 60 mM potassium chloride (%KCl). Welch corrected unpaired t-test, *, p < 0.05 vs. tyramine; ^, p < 0.05 vs. synephrine.

that tyramine and octopamine acts directly on TAARs to cause in endothelium-denuded rat aortas15,16 and pig coronary arteries17, but no selective antagonists were available to investigate this hypothesis. Te presence of specifc tyramine receptors had previously been proposed to explain rat aortic responses to syn- ephrine that were resistant to conventional receptor antagonists8. Tus, there is some indirect evidence to suggest that tyramine and synephrine may induce some vascular responses via specifc receptors. Te aim of the present study was to investigate the actions and mechanisms of action of the trace amines on the mesenteric and coronary artery using a porcine model. Te mesenteric artery is densely innervated by the sympathetic nerves, and is a major regulator of blood fow to the gastrointestinal system during physiological stress18. Blood fow to the intestine can be drastically reduced (>80%) during which results in the shunt- ing of blood to the skeletal and cardiac muscle19. In contrast, during physiological stress the nervous system ensures an increased blood fow to cardiac muscle via the coronary circulation20,21. Tus, in this study, the relative potencies of the trace amines tyramine, octopamine and synephrine were investigated on the vascular tone of mesenteric artery and the coronary artery and the mechanisms of action for each amine in the two func- tionally very diferent arteries determined. Te role of TAAR-1 in responses was also examined using EPPTB (RO-5212773), a recently developed selective antagonist for TAAR-122. Results Responses of the porcine mesenteric artery. All three trace amines produced concentration-depend- ent contractions of the porcine arterial rings (Fig. 2). Tyramine and synephrine produced similar maximum responses, whilst those to octopamine were signifcantly greater (unpaired Welch’s t-test, p < 0.05). All three trace amines had similar potencies that were not signifcantly diferent (pEC50 ranging from 3.25–3.91). Te endog- enous amine, noradrenaline produced greater contractions than the three trace amines and was also the most potent drug tested (Fig. 2, Table 1).

Responses of the porcine coronary artery. All three trace amines produced concentration-dependent contractions of porcine coronary arterial rings (Fig. 3). Octopamine and synephrine produced similar maximum responses, whilst those to tyramine were greater, the diference between tyramine and octopamine being statis- tically signifcant (p < 0.05). Te potencies of all three amines were similar and ranged from 3.30–3.88 (Table 2).

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 2 www.nature.com/scientificreports/ www.nature.com/scientificreports

Controls Treatment and interventions Max contraction (%KCl) Potency (pEC50) Max contraction (%KCl) Potency (pEC50) Sample number (n) Tyramine Tyramine pre-treatment (3 mM) 30.0 ± 3.9 3.91 ± 0.14 2.9 ± 0.7* 2.87 ± 0.06* 6 Propranolol (1 µM) 2.9 ± 0.7 2.49 ± 0.19 2.9 ± 0.6 2.88 ± 0.27 6 Prazosin (1 µM) 2.9 ± 0.7 2.49 ± 0.19 Abolished* Abolished* 6 L-NNA (100 µM) 4.2 ± 2.7 2.50 ± 0.23 6.5 ± 3.0 2.81 ± 0.56 4 Synephrine Tyramine pre-treatment (3 mM) 25.6 ± 4.8 3.81 ± 0.12 10.9 ± 2.6* 3.76 ± 0.26 6 Propranolol (1 µM) 12.5 ± 2.9 3.90 ± 0.51 16.1 ± 3.8 3.45 ± 0.41 6 Prazosin (1 µM) 9.6 ± 2.6 3.81 ± 0.56 Abolished* Abolished* 5 L-NNA (100 µM) 19.2 ± 10.1 4.06 ± 0.68 17.4 ± 9.8 2.29 ± 1.19 4 Octopamine Tyramine pre-treatment (3 mM) 63.4 ± 6.3 3.25 ± 0.12 19.9 ± 2.3* 3.60 ± 0.21 5 Propranolol (1 µM) 17.1 ± 3.6 3.80 ± 0.44 15.4 ± 4.4 3.46 ± 0.52 5 Prazosin (1 µM) 14.0 ± 2.8 3.80 ± 0.45 Abolished* Abolished* 6 L-NNA (100 µM) 16.7 ± 9.3 3.80 ± 0.44 12.8 ± 6.0 3.55 ± 0.41 6 Noradrenaline Tyramine pre-treatment (3 mM) 85.7 ± 12.2 5.42 ± 0.21 85.9 ± 6.5 5.22 ± 0.13 5 Propranolol (1 µM) 81.9 ± 5.9 5.92 ± 0.14 91.4 ± 4.4 5.65 ± 0.09 5 Prazosin (1 µM) 87.7 ± 4.7 5.92 ± 0.14 81.9 ± 5.2 5.10 ± 0.08* 5

Table 1. Efects of antagonists on the maximum responses and potency values of p-synephrine, octopamine, tyramine, noradrenaline and in porcine mesenteric arteries. Paired Student’s t-test *p < 0.05 vs control.

Te endogenous amine noradrenaline failed to produce contraction and only relaxations were observed (Fig. 3). Tese relaxations were converted to contractions in the presence of the β-adrenoceptor antagonist propranolol (1 µM). Te maximum contractile responses to noradrenaline (33.4 ± 2.7% of the response to potassium) were signifcantly greater than those to octopamine (20.3 ± 4.6%, p < 0.05) and synephrine (21.1 ± 7.3%, p < 0.05), but not tyramine (32.3 ± 4.5%). In the presence of both propranolol (1 µM) and the α1-adrenoceptor antagonist prazosin (1 µM) responses to noradrenaline were abolished completely.

The role of endogenous noradrenaline release in responses in mesenteric artery. In mes- enteric arteries, tyramine pre-treatment for 30 minutes nearly abolished subsequent responses to tyramine with a > 90% reduction in maximal contraction and a rightward shif of curves (pEC50 value 2.87 ± 0.06, n = 6; Student’s t-test p < 0.05) (Fig. 4). In tissues depleted of endogenous noradrenaline, the maximal con- tractions to synephrine were reduced by 58% and maximal responses to octopamine were halved. However, the 30-minute tyramine pre-treatment did not afect the potencies (pEC50 values) of either synephrine or octopamine (Fig. 4). Te contractile responses to exogenous noradrenaline were not signifcantly afected by tyramine pre-treatment (Table 1). To examine whether responses to any of the trace amines involved the release of nitric oxide, tissues were incubated with the nitric oxide synthase inhibitor, L-NNA (100 µM). None of the responses to the trace amines was afected by the removal of nitric oxide (Table 1).

The role of endogenous noradrenaline release in responses in coronary artery. Tyramine pre-treatment did not afect the potencies or maximal contractions to either tyramine itself or synephrine (Fig. 4, Table 2). Noradrenaline depletion did appear to reduce responses to octopamine by about 50%, but the change was not statistically signifcant (Fig. 4). None of the responses to the amines was afected by the removal of nitric oxide with L-NNA (Table 2).

Role of direct adrenoceptor stimulation in trace amine-induced vasoconstriction in mesenteric artery. In noradrenaline-depleted mesenteric artery rings the presence of the α1-adrenoceptor antagonist pra- zosin (1 µM) abolished the responses to all three trace amines (Fig. 5, Table 1). Prazosin caused a rightward shif of concentration-response curves to noradrenaline (P < 0.05) without afecting maximum responses (Table 1). Neither maximum contractile responses nor potencies for any of the trace amines or noradrenaline were changed in the presence of propranolol (1 µM) (Table 1).

Role of direct adrenoceptor stimulation in trace amine-induced vasoconstriction in coronary artery. In noradrenaline-depleted coronary artery rings the presence of prazosin (1 µM) reduced maximum contractions produced by tyramine, octopamine and synephrine by about half without signifcantly afecting agonist potency (Fig. 5, Table 2). Contractile responses to synephrine were the most signifcantly afected by pra- zosin, but responses were again not completely abolished. Neither maximum contractile responses nor potencies for any of the amines changed signifcantly in the presence of propranolol (1 µM) (Table 2).

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 3 www.nature.com/scientificreports/ www.nature.com/scientificreports

Noradrenaline (Control) 40 Tyramine 40 * Octopamine Propranonlol (1 µM) * ) 30 * Synephrine Prazosin (1 µM) *

30 KCl) 20 * %KCl (% * n( 10 20 * * * * * * * 0

10 -10 Contractio Contracti on -20 0 -8 -7 -6 -5 -4 -6 -5 -4 -3 -2 log[Noradrenaline] log[Trace amine]

Figure 3. (Lef) Concentration-response curves of the porcine coronary artery to tyramine, octopamine and synephrine (n = 6–8). (Right) Concentration-response curves to noradrenaline in the absence of antagonists and in the presence of the β-adrenoceptor antagonist propranolol (1 M) and in the combined presence of propranolol (1 µM) and the α1-adrenoceptor antagonist prazosin (1 µM). Responses are expressed as a percentage of the contractile response to potassium chloride (60 mM). Student’s paired parametric t-test *p < 0.05 vs. the relaxation before addition of antagonists.

Controls Treatment Trace amine and interventions Max contraction (%KCl) Potency (pEC50) Max contraction (%KCl) Potency (pEC50) Sample number (n) Tyramine Tyramine pre-treatment (3 mM) 32.3 ± 4.5 3.30 ± 0.13 34.12 ± 6.5 3.37 ± 0.12 8 Propranolol (1 µM) 24.1 ± 3.8 3.27 ± 0.12 22.3 ± 2.9 3.23 ± 0.11 5 Prazosin (1 µM) 39.3 ± 7.5 3.51 ± 0.17 24.3 ± 5.9** 3.26 ± 0.21 6 EPPTB (100 nM) 22.8 ± 7.5 3.03 ± 0.31 31.4 ± 6.7 2.94 ± 0.14 6 EPPTB (100 µM) 13.4 ± 2.1 3.13 ± 0.09 4.0 ± 1.1** 3.10 ± 0.18 7 L-NNA (100 µM) 34.6 ± 4.5 3.42 ± 0.20 41.6 ± 4.1 3.27 ± 0.14 6 Octopamine Tyramine pre-treatment (3 mM) 20.3 ± 4.6 4.04 ± 0.43 13.7 ± 3.9 3.58 ± 0.29 6 Propranolol (1 µM) 12.6 ± 4.5 3.85 ± 0.68 20.7 ± 6.3 3.35 ± 0.34 5 Prazosin (1 µM) 13.7 ± 3.9 3.55 ± 0.27 5.9 ± 2.6* 3.15 ± 0.24* 6 EPPTB (100 nM) 6.5 ± 2.0 3.31 ± 0.88 8.1 ± 2.9 2.94 ± 0.25 5 EPPTB (100 µM) 16.5 ± 2.9 3.43 ± 0.18 7.8 ± 2.5* 3.04 ± 0.20 6 L-NNA (100 µM) 18.8 ± 6.0 3.13 ± 0.49 17.6 ± 1.5 3.27 ± 0.13 5 Synephrine Tyramine pre-treatment (3 mM) 21.1 ± 7.3 3.50 ± 0.26 13.1 ± 3.4 3.55 ± 0.19 6 Propranolol (1 µM) 10.5 ± 3.0 3.74 ± 0.44 16.9 ± 0.3 2.59 ± 0.29 7 Prazosin (1 µM) 22.5 ± 6.7 3.21 ± 0.80 6.0 ± 2.1* 3.70 ± 0.35 6 EPPTB (100 nM) 25.1 ± 7.2 3.42 ± 0.43 27.0 ± 5.1 3.54 ± 0.33 6 EPPTB (100 µM) 12.8 ± 4.3 3.82 ± 0.42 3.9 ± 1.3* 3.83 ± 0.36 8 L-NNA (100 µM) 34.4 ± 6.5 3.29 ± 0.28 24.8 ± 4.5 3.60 ± 0.31 6 Noradrenaline Propranolol (1 µM) −16.9 ± 1.5 6.98 ± 0.48 33.4 ± 2.7** 5.40 ± 0.15* 6 Propranolol + Prazosin (1 µM) −16.9 ± 1.5 6.98 ± 0.48 4.0 ± 1.5** 5.29 ± 0.65* 6

Table 2. Efects of antagonists on the maximum responses and potency values of tyramine, octopamine, synephrine and noradrenaline in porcine coronary arteries. Paired parametric Student’s t-test *vs. controls p < 0.05. **vs. controls p < 0.01.

Role of Trace amine-associated receptor 1 (TAAR-1) in trace amine-induced vasoconstriction of coronary artery. In noradrenaline-depleted tissues,the possible involvement of the TAAR-1 receptor in mediating responses was investigated using the selective TAAR-1 antagonist EPPTB (Fig. 6, Table 2). Te efects of this antagonist were dependent on the concentration of the antagonist. At the lower concentration (100 nM), EPPTB did not reduce responses to any of the trace amines, and for tyramine, responses were greater in the pres- ence of this antagonist, although the change was not statistically signifcant. Te higher concentration of EPPTB (100 µM) did not afect pEC50 values for any of the amines but reduced maximum responses to both tyramine and synephrine by 70%, whilst responses to octopamine were reduced by 53% (Fig. 6, Table 2).

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 4 www.nature.com/scientificreports/ www.nature.com/scientificreports

40 A 80 B 40 C ) ) )

Cl Control Cl Cl 30 Noradrenaline depleted 60 30 %K %K %K n( n( n( 20 40 20 ** * * * * 10 20 10 * ont ractio ont ractio ont ractio

C ** C C * * 0 0 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 Log [Tyramine] log [Octopamine] log [Synephrine]

50 D 35 E 35 F ) ) ) 30 30 Cl Cl Cl 40 25 25 %K %K %K 30 n( n( n( 20 20 15 15 actio 20 actio actio 10 10 ontr ontr ontr 10 C C C 5 5 0 0 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 Log [Tyramine] log [Octopamine] log [Synephrine]

Figure 4. Concentration-response curves to tyramine (A,D), octopamine (B,E) and synephrine (C,F) in control tissues (○), mesenteric (A–C) and coronary arteries (D–F) previously depleted of neuronal NA (▪) using prolonged contact (60 mins) with a high concentration (3 mM) of tyramine. Data are means ± sem values from 6 to 24 experiments, expressed as a percentage of the contractile response to potassium chloride (60 mM). Student’s paired parametric t-test *p < 0.05 vs. control.

5 A 25 B 25 C ) ) ) Control Cl Cl Cl 4 20 20 Prazosin %K %K %K 15 15

3 n( n( n( io 2 10 10

ontract 5 5

1 Contractio Contractio C * * * * * * * * * 0 0 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 log [Tyramine] log [Octopamine] log [Synephrine]

50 30 30 D E F ) ) )

Cl 40 Control Cl Cl Prazosin (1 µM) %K %K 20 %K 20 30 ** n( * n( n( *

actio 20 actio actio 10 10 * * * * * * 10 * * Contr Contr Contr * * * * 0 0 * 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 Log [Tyramine] Log [Octopamine] Log [Synephrine]

Figure 5. Concentration-response curves to trace amines in the absence (○) and presence of prazosin (1 µM, ■) in NA-depleted mesenteric arteries (A–C); and NA-depleted coronary arterial rings (D–F). Data are means ± sem (n = 5–7) values expressed as a percentage of the contractile response to potassium chloride (60 mM). Student’s paired parametric t-test *p < 0.05 vs control.

Discussion All three trace amines contracted the mesenteric artery but were weaker and less potent than the noradrenaline. Te potency order for the amines was noradrenaline > tyramine = synephrine = octopamine where the diferences between the three trace amines were not signifcantly diferent. Previously vasoconstrictor efects of tyramine on canine mesenteric arteries23,24 and octopamine on rat mesenteric arterioles25 have been observed, but the vascular efects of synephrine are unclear. Huang et al. (1995) reported that synephrine was incapable of producing contractions of isolated rat mesenteric arteries26, but in our study synephrine produced concentration-dependent vasoconstrictions of the porcine mesenteric artery, suggesting that species diferences may occur. In the same study by Huang and colleagues, a Citrus aurantium extract caused dose-dependent

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 5 www.nature.com/scientificreports/ www.nature.com/scientificreports

40 20 40 A B C ) ) ) Control Cl Cl Cl 30 EPPTB (100nM) 15 30 %K %K %K n( n( n(

io 20 io 10 io 20

10 5 10 ontract ontract ontract C C C

0 0 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 Log [Tyramine] Log [Octopamine] Log [Synephrine]

20 20 20 D E F ) )

Cl Control Cl 15 EPPTB (100µM) 15 15 %K %K (%KCl)

n( n( * * on io 10 io 10 10

** ** * 5 5 * ** ** 5

Contract Contract *

* Contracti * * * * 0 0 0 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 -6 -5 -4 -3 -2 Log [Tyramine] Log [Octopamine] Log [Synephrine]

Figure 6. Concentration-response curves to tyramine (A,D), octopamine (B,E) and synephrine (C,F) in the absence (○) and presence of the TAAR-1 antagonist, EPPTB at 100 nM (■;A–C) or 100 µM (▲;D–F) in porcine coronary arterial rings. Data are means ± sem from 5 to 8 separate experiments, expressed as a percentage of the contractile response to potassium chloride (60 mM). Student’s paired parametric t-test *p < 0.05 vs. control, **p < 0.01 vs. control.

vasocontractions of the rat mesenteric artery with a maximum value of 27 ± 3% (% 60 mM KCl), a value similar to the maximum contractions observed in the responses to synephrine in this current study. In contrast to our experiments on isolated mesenteric arteries, Anwar and colleagues showed a vasodilatory response in perfused rat mesenteric beds6. Te discrepancy between the reports may be a result of diferences in perivascular nerve distribution, as there is a greater density in non-adrenergic non-cholinergic nerves 2nd and 3rd branches of mes- enteric arteries27. Tus diferences between the Anwar study and our study may be explained by diferences in species (rat vs pig) of type of vessel examined (perfused vascular bed vs larger artery). On the coronary artery, all three trace amines were partial and contracted the tissues with potencies similar to those seen on the mesenteric artery (ie. pEC50 range 3.3–4.0). Te observations for tyramine were consistent with previous studies on the porcine coronary artery17 and the rat aorta8,9,15. However, unlike the mesenteric artery, the lef anterior coronary artery contains a high proportion of β2-adrenoceptors and relatively 28 few α1-adrenoceptors and noradrenaline administered to control arteries caused vasorelaxation, which would be consistent with predominant β2-adrenoceptor activation. In the presence of the β-adrenoceptor antagonist, propranolol, the responses to noradrenaline were reversed to vasoconstriction. Te constriction of the coronary smooth muscle in response to noradrenaline was caused by α1-adrenoceptors as evidenced by the abolition of responses in the presence of the α1-adrenoceptor antagonist prazosin. In contrast to noradrenaline, propranolol did not afect the responses to the trace amines, and even in the presence of prazosin, these amines could produce coronary artery vasoconstriction suggesting a non-adrenoceptor . To further investigate the mechanisms involved in the vasoconstriction induced by the trace amines, the con- tribution of indirect actions and direct adrenoceptor activity were examined. Te inhibition of NO synthase did not signifcantly afect responses to any of the amines indicating they did not stimulate release of NO. However, depletion of sympathetic stores of noradrenaline did have an efect and has been shown to greatly reduce the responses of tissues to indirectly acting sympathomimetic drugs29,30. Tyramine was frst described as an indi- rectly acting by Burn and Rand (1958), who showed that the depletion of endogenous noradrenaline stores, completely abolished the efects of tyramine but not directly acting amines31. Te mesenteric vasculature contains a dense neuronal network that would redirect blood fow away from the intestine during sympathetic nervous system activation18. Te incubation of porcine mesenteric arteries with a high concentration (3 mM) of tyramine for 30 minutes almost abolished subsequent responses of tissues to tyramine, indicating that stores of noradrenaline were depleted. Responses to exogenous noradrenaline were not afected by prior depletion of noradrenaline stores, demonstrating direct activation of α-adrenoceptors by noradrenaline caused vasoconstriction. In contrast, the responses to octopamine and synephrine were halved in noradrenaline-depleted tissues, suggesting that, like tyramine, the release of endogenous noradrenaline from sympathetic nerves contributes, at least partially, to the responses of these mesenteric vessels to these trace amines. Diferent results were obtained in the coronary arteries where the three trace amines continued to elicit sig- nifcant responses afer incubations to deplete endogenous noradrenaline stores. Responses to tyramine were almost identical in control and noradrenaline-depletion tissues, however responses to octopamine and syneph- rine appeared to be reduced slightly but the changes were small and not statistically signifcant. Tese results suggest that in the coronary artery, unlike the mesenteric artery, the release of endogenous noradrenaline from sympathetic nerves does not contribute signifcantly to the responses of coronary vessels to trace amines. Tis

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 6 www.nature.com/scientificreports/ www.nature.com/scientificreports

conclusion was further supported by the fnding that the response of the coronary artery to noradrenaline in the absence of antagonists was relaxation, making it impossible for the release of endogenous noradrenaline to cause the contractile responses to the trace amines. In mesenteric arteries the small contractile responses to the trace amines remaining afer noradrenaline depletion were not afected by the β-adrenoceptor antagonist, propranolol, but were abolished in the presence of prazosin. Tis indicates that the amines lacked signifcant efects on vascular β-adrenoceptors in this and that the residual contractile efects afer noradrenaline depletion were mediated via α1-adrenoceptor stim- ulation. It has previously been reported that responses of the coronary artery to the trace amines, tyramine and β-phenylethylamine (β-PEA) were not afected by either prazosin or propranolol17, suggesting a lack of adreno- ceptor involvement in responses. However, the coronary responses to synephrine and octopamine were not explored. In the present study, the maximum contractions of coronary arteries to all three trace amines were signifcantly reduced by prazosin indicating a weak action at the α1-adrenoceptors of this tis- sue. Tese results are consistent with previous data using rat aorta where octopamine and synephrine activated 7,9,15 α1-adrenoceptors to cause contraction . However, the residual vasoconstrictions to these three amines in the presence of prazosin suggests the involvement of another non-. Other monoamines such as 5-HT, and are known to cause constriction of pig coronary arteries, but blockade of their respective receptors does not afect responses to tyramine or β-PEA17. Tis prompted the authors to propose that these amines were acting on TAARs although an antagonist was not available at the time to test this hypothesis. Te physiological efects of trace amine-associated receptors were frst identifed in the process of olfaction, but these receptors have since been identifed in tissues throughout the body. Te TAAR-1 subtype is unusual as it is not involved in olfaction but is involved in and is therefore a potential target for drug development in the treatment of psychiatric and neurodegenerative disorders32. Stimulation of TAAR-1 causes activation of the G-protein subunit (Gs) and subsequent increase in cellular cAMP concentration12. Te selective competitive antagonist for TAAR-1, EPPTB (RO-5212773) was frst described by33. It has a high afnity for the mouse TAAR-1 (Ki = 0.9 nM) and was employed in the present study at a concentration of 100 nM. Te maximum responses to tyramine, but not octopamine or synephrine, appeared to be increased in the presence of 100 nM EPPTB although the efect was not statistically signifcant. Had tyramine activated TAAR-1 receptors enhanced responses would be expected since the receptor normally inhibits contraction. Tis would ft with the known activation of following activation of this receptor and the inhibition of contraction by intracellular cAMP. It has been reported that tyramine is more potent at TAAR-1 than octopamine or syneph- rine13,14. However, the efects of EPPTB are known to be species-dependent33 since the drug exhibited a high afn- ity for the mouse variant of TAAR-1 (Ki = 0.9 nM) but a much lower afnity for the rat (Ki = 942 nM) and human TAAR-1 (Ki > 5 µM). Te afnity of EPPTB at the porcine receptor is unknown, but we examined a higher con- centration of the antagonist that would block responses mediated via the low-afnity variant of this receptor. In the presence of the higher concentration of EPPTB (100 µM), contractile responses to all three trace amines were reduced by more than half. Whether this represents an action of these trace amines at the low afnity TAAR-1 or a non-selective action at other TAAR receptors at higher drug concentrations is not known. Unfortunately, further studies will have to wait until selective drugs at other TAAR subtypes have been developed. Conclusion Tis study has shown that the trace amines, (tyramine, octopamine and synephrine) can induce vasoconstriction by a variety of complex mechanisms including direct α-adrenoceptor activation, indirect actions via release of endogenous noradrenaline and actions at another receptor yet to be identifed, with the mechanism depending on the blood vessel. On the mesenteric artery, the trace amines induced contractions via an indirect sympathom- imetic action and also a direct α1-adrenoceptor agonist mechanism. In the coronary artery, the amines appeared to have almost no indirect sympathomimetic actions but do elicit contractions via weak direct agonist actions on α1-adrenoceptors and possibly an action on TAAR receptors other than the TAAR-1 subtype. Methods and Materials Tissue preparation. Hearts and gastrointestinal tracts from 5-month-old female pigs were obtained from the local abattoir and transported in ice-cold Krebs-bicarbonate solution to the laboratory. Te lef anterior descending coronary artery or the inferior mesenteric artery were isolated and 3 mm length rings were suspended between stainless-steel hooks and stationary supports in 8 mL organ bath (EZ-baths, Global Towns, CA) contain- ing Krebs-bicarbonate solution (composition in mM: NaCl 118, NaHCO3 25, glucose 11.7, MgSO4 2.4, KH2PO4 1.2, KCl 1.2, CaCl2 2.5) maintained at 37 °C and continuously oxygenated with 5% CO2 in oxygen. Te rings were mounted under a resting tension of 5 g and the tension developed by the circular smooth muscle in response to the addition of drugs was measured using isometric force transducers coupled to a PowerLab computer system (AD Instruments, Castle Hill, Australia).

Experimental protocol. Afer equilibrating the tissues for 30 minutes, cumulative concentration-response curves to either noradrenaline, phenylephrine, tyramine, synephrine or octopamine were generated. Afer wash- out, responses to the amines were repeated and responses to KCl (60 mM) were obtained at the end of the exper- iment. Responses were expressed as a percentage of the response to 60 mM KCl. To investigate the possible indirect sympathomimetic action of these amines; the tissues were incubated with a high concentration of tyramine (3 mM) for 30 minutes to deplete pre-synaptic noradrenaline (NA) stores. Afer washout, responses to noradrenaline, tyramine, synephrine or octopamine were again obtained. To investigate the mechanisms involved in the functional actions of these amines; α1- or β-adrenoceptor antagonists (prazosin or propranolol respectively; both 1 µM), trace amine-associated receptor 1 antagonist (EPPTB; 100 nM and 100 µM) or the nitric oxide synthase inhibitor, L-Nω-Nitroarginine (L-NNA; 100 µM) were

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 7 www.nature.com/scientificreports/ www.nature.com/scientificreports

added 30 minutes before the addition of agonists. Control experiments without the addition of the antagonist were also performed, but repeated curves to the amines were not signifcantly diferent and correction of the second test curve in the presence of the antagonist was not required.

Statistical analysis. Concentration-response curves were analysed using PRISM 8 (GraphPad Sofware, San Diego, USA). Te data are represented as mean ± s.e. mean with n indicating the number of animals from which arterial rings were obtained. Comparison of the diferent concentration-response curves was performed using paired Student’s t-test when comparing 2 normally distributed groups (e.g. pEC50 in the absence v.s. presence of antagonists), or a Welch corrected unpaired Student’s t-test applied to the normalised responses which may have unequal variances. Statistical signifcance was demonstrated by a p-value of less than 0.05. Potency was expressed as the pEC50 value (-log EC50, which was the molar concentration producing a response 50% of the maximum efect).

Drug and chemicals. (+)Noradrenaline-bitartrate, (±)p-synephrine, (±)p-octopamine hydrochloride, (±)p-tyramine, (+)-phenylephrine hydrochloride, prazosin hydrochloride, propranolol hydrochloride, EPPTB (RO-5212773, N-(3-Ethoxyphenyl)-4-(1-pyrrolidinyl)-3-(trifuoromethyl)benzamide) and Nω- nitro-L-arginine (L-NNA) were obtained from Sigma-Aldrich (Sydney, Australia). All drugs except RO-5212773 were dissolved in distilled water and then diluted with Krebs-bicarbonate solution. RO-5212773 was dissolved in 1% dimethyl sulfoxide (DMSO) prior to dilution with Krebs-bicarbonate solution. References 1. D’Andrea, G. et al. Abnormal in chronic cluster headache. Cephalalgia 37, 148–153, https://doi. org/10.1177/0333102416640502 (2017). 2. Pellati, F. & Benvenuti, S. Chromatographic and electrophoretic methods for the analysis of [corrected] in Citrus aurantium. J Chromatogr A 1161, 71–88, https://doi.org/10.1016/j.chroma.2007.05.097 (2007). 3. Rossato, L. G., Costa, V. M., Limberger, R. P., Bastos Mde, L. & Remiao, F. Synephrine: from trace concentrations to massive consumption in weight-loss. Food Chem Toxicol 49, 8–16, https://doi.org/10.1016/j.fct.2010.11.007 (2011). 4. Eudy, A. E. et al. Efcacy and safety of ingredients found in preworkout supplements. Am J Health Syst Pharm 70, 577–588, https:// doi.org/10.2146/ajhp120118 (2013). 5. Broadley, K. J. Te vascular efects of trace amines and . Pharmacol Ter 125, 363–375, https://doi.org/10.1016/j. pharmthera.2009.11.005 (2010). 6. Anwar, M. A., Ford, W. R., Broadley, K. J. & Herbert, A. A. Vasoconstrictor and vasodilator responses to of rat-isolated perfused mesentery: comparison with tyramine and beta-phenylethylamine. Br J Pharmacol 165, 2191–2202, https://doi. org/10.1111/j.1476-5381.2011.01706.x (2012). 7. Brown, C. M. et al. Activities of octopamine and synephrine stereoisomers on alpha-adrenoceptors. Br J Pharmacol 93, 417–429 (1988). 8. Varma, D. R., Deng, X. F., Chemtob, S., Nantel, F. & Bouvier, M. Characterization of the vasorelaxant activity of tyramine and other phenylethylamines in rat aorta. Can J Physiol Pharmacol 73, 742–746 (1995). 9. Hibino, T., Yuzurihara, M., Kase, Y. & Takeda, A. synephrine, a component of Evodiae Fructus, constricts isolated rat aorta via adrenergic and serotonergic receptors. J Pharmacol Sci 111, 73–81 (2009). 10. Kleinau, G. et al. Diferential modulation of Beta-adrenergic receptor signaling by trace amine-associated receptor 1 agonists. PLoS One 6, e27073, https://doi.org/10.1371/journal.pone.0027073 (2011). 11. Jordan, R., Midgley, J. M., Tonoor, C. M. & Williams, C. M. Beta-adrenergic activities of octopamine and synephrine stereoisomers on guinea-pig atria and . Te Journal of pharmacy and pharmacology 39, 752–754 (1987). 12. Bunzow, J. R. et al. , 3,4-methylenedioxymethamphetamine, lysergic acid diethylamide, and metabolites of the are agonists of a rat trace amine receptor. Mol Pharmacol 60, 1181–1188 (2001). 13. Borowsky, B. et al. Trace amines: identifcation of a family of mammalian -coupled receptors. Proc Natl Acad Sci USA 98, 8966–8971, https://doi.org/10.1073/pnas.151105198 (2001). 14. Wainscott, D. B. et al. Pharmacologic characterization of the cloned human trace amine-associated receptor1 (TAAR1) and evidence for species diferences with the rat TAAR1. Te Journal of pharmacology and experimental therapeutics 320, 475–485, https://doi. org/10.1124/jpet.106.112532 (2007). 15. Broadley, K. J., Fehler, M., Ford, W. R. & Kidd, E. J. Functional evaluation of the receptors mediating vasoconstriction of rat aorta by trace amines and amphetamines. Eur J Pharmacol 715, 370–380, https://doi.org/10.1016/j.ejphar.2013.04.034 (2013). 16. Fehler, M., Broadley, K. J., Ford, W. R. & Kidd, E. J. Identifcation of trace-amine-associated receptors (TAAR) in the rat aorta and their role in vasoconstriction by beta-phenylethylamine. Naunyn Schmiedebergs Arch Pharmacol 382, 385–398, https://doi. org/10.1007/s00210-010-0554-1 (2010). 17. Herbert, A. A., Kidd, E. J. & Broadley, K. J. Dietary trace amine-dependent vasoconstriction in porcine coronary artery. Br J Pharmacol 155, 525–534, https://doi.org/10.1038/bjp.2008.286 (2008). 18. Ceppa, E. P., Fuh, K. C. & Bulkley, G. B. Mesenteric hemodynamic response to circulatory . Curr Opin Crit Care 9, 127–132 (2003). 19. Clausen, J. P. Efect of physical training on cardiovascular adjustments to exercise in man. Physiol Rev 57, 779–815, https://doi. org/10.1152/physrev.1977.57.4.779 (1977). 20. Santulli, G. & Iaccarino, G. Adrenergic signaling in and cardiovascular aging. Maturitas 93, 65–72, https://doi. org/10.1016/j.maturitas.2016.03.022 (2016). 21. Goodwill, A. G., Dick, G. M., Kiel, A. M. & Tune, J. D. Regulation of Coronary Blood Flow. Compr Physiol 7, 321–382, https://doi. org/10.1002/cphy.c160016 (2017). 22. Stalder, H., Hoener, M. C. & Norcross, R. D. Selective antagonists of mouse trace amine-associated receptor 1 (mTAAR1): discovery of EPPTB (RO5212773). Bioorg Med Chem Lett 21, 1227–1231, https://doi.org/10.1016/j.bmcl.2010.12.075 (2011). 23. Toda, N., Hayashi, S. & Hattori, K. Analysis of the efect of tyramine and in isolated canine cerebral and mesenteric arteries. Te Journal of pharmacology and experimental therapeutics 205, 382–391 (1978). 24. Muramatsu, I. Te efect of on sympathetic, purinergic neurotransmission in the isolated mesenteric artery of the dog: a pharmacological study. Br J Pharmacol 91, 467–474 (1987). 25. Altura, B. M. Pharmacological efects of alpha-, alpha-methylnorepinephrine, and octopamine on rat arteriolar, arterial, and terminal vascular smooth. Circ Res 36, 233–240 (1975). 26. Huang, Y. T. et al. Fructus aurantii reduced portal pressure in portal hypertensive rats. Life sciences 57, 2011–2020 (1995). 27. Yokomizo, A., Takatori, S., Hashikawa-Hobara, N., Goda, M. & Kawasaki, H. Characterization of Perivascular Nerve Distribution in Rat Mesenteric Small Arteries. Biol Pharm Bull 38, 1757–1764, https://doi.org/10.1248/bpb.b15-00461 (2015). 28. Corr, L. & Burnstock, G. Vasodilator response of coronary smooth muscle to the sympathetic co-transmitters noradrenaline and 5′-triphosphate. Br J Pharmacol 104, 337–342 (1991).

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 8 www.nature.com/scientificreports/ www.nature.com/scientificreports

29. Grobecker, H., Roizen, M. F. & Kopin, I. J. Effect of tyramine and on dopamine-beta-hydroxylase activity and norepinephrine concentrations in vesicular fraction of the heart and plasma of rats. Life sciences 20, 1009–1015 (1977). 30. Toa, N. B., Wooten, G. F., Axelrod, J. & Kopin, I. J. On the mechanism of release of norepinephrine from sympathetic nerves induced by depolarizing agents and sympathomimetic drugs. Mol Pharmacol 11, 10–18 (1975). 31. Burn, J. H. & Rand, M. J. Te action of sympathomimetic amines in animals treated with reserpine. J Physiol 144, 314–336 (1958). 32. Lindemann, L. et al. Trace amine-associated receptors form structurally and functionally distinct subfamilies of novel G protein- coupled receptors. Genomics 85, 372–385, https://doi.org/10.1016/j.ygeno.2004.11.010 (2005). 33. Bradaia, A. et al. Te selective antagonist EPPTB reveals TAAR1-mediated regulatory mechanisms in of the mesolimbic system. Proc Natl Acad Sci USA 106, 20081–20086, https://doi.org/10.1073/pnas.0906522106 (2009). Author Contributions Andy Hsien Weh Koh – study design, data collection, analysis & interpretation, drafing of manuscript; Russ Chess-Williams – study design, data analysis & interpretation, critical revision of manuscript; Anna Lohning - study design, data analysis & interpretation, critical revision of manuscript, fnal approval. Additional Information Competing Interests: Te authors declare no competing interests. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. Te images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© Te Author(s) 2019

Scientific Reports | (2019)9:10925 | https://doi.org/10.1038/s41598-019-46627-5 9